Given ten observations x 1 , x 2 , … , x 10 x_1, x_2, \ldots, x_{10} x 1 , x 2 , … , x 10 with:
∑ i = 1 10 ( x i − 2 ) = 30 , ∑ i = 1 10 ( x i − β ) 2 = 98 , β > 2 , \sum_{i=1}^{10} (x_i - 2) = 30, \quad \sum_{i=1}^{10} (x_i - \beta)^2 = 98, \quad \beta > 2, ∑ i = 1 10 ( x i − 2 ) = 30 , ∑ i = 1 10 ( x i − β ) 2 = 98 , β > 2 ,
and variance 4 5 \frac{4}{5} 5 4 . We need to find β μ σ 2 \frac{\beta \mu}{\sigma^2} σ 2 β μ , where μ \mu μ and σ 2 \sigma^2 σ 2 are the mean and variance of the new data set y i = 2 ( x i − 1 ) + 4 β y_i = 2(x_i - 1) + 4\beta y i = 2 ( x i − 1 ) + 4 β .
First, recall that variance is given by 1 n ∑ ( x i − x ˉ ) 2 \frac{1}{n} \sum (x_i - \bar{x})^2 n 1 ∑ ( x i − x ˉ ) 2 . With n = 10 n = 10 n = 10 and variance 4 5 \frac{4}{5} 5 4 :
1 10 ∑ i = 1 10 ( x i − x ˉ ) 2 = 4 5 \frac{1}{10} \sum_{i=1}^{10} (x_i - \bar{x})^2 = \frac{4}{5} 10 1 ∑ i = 1 10 ( x i − x ˉ ) 2 = 5 4
Multiplying both sides by 10:
∑ i = 1 10 ( x i − x ˉ ) 2 = 8 (Equation 1) \sum_{i=1}^{10} (x_i - \bar{x})^2 = 8 \quad \text{(Equation 1)} ∑ i = 1 10 ( x i − x ˉ ) 2 = 8 (Equation 1)
From ∑ i = 1 10 ( x i − 2 ) = 30 \sum_{i=1}^{10} (x_i - 2) = 30 ∑ i = 1 10 ( x i − 2 ) = 30 :
∑ i = 1 10 ( x i − 2 ) = ∑ x i − 20 = 30 \sum_{i=1}^{10} (x_i - 2) = \sum x_i - 20 = 30 ∑ i = 1 10 ( x i − 2 ) = ∑ x i − 20 = 30
So:
∑ x i = 50 (Equation 2) \sum x_i = 50 \quad \text{(Equation 2)} ∑ x i = 50 (Equation 2)
The mean x ˉ \bar{x} x ˉ is:
x ˉ = 50 10 = 5 (Equation 3) \bar{x} = \frac{50}{10} = 5 \quad \text{(Equation 3)} x ˉ = 10 50 = 5 (Equation 3)
Using Equation 1 and x ˉ = 5 \bar{x} = 5 x ˉ = 5 :
∑ i = 1 10 ( x i − 5 ) 2 = ∑ ( x i 2 − 10 x i + 25 ) = ∑ x i 2 − 10 ∑ x i + 250 = 8 \sum_{i=1}^{10} (x_i - 5)^2 = \sum (x_i^2 - 10x_i + 25) = \sum x_i^2 - 10 \sum x_i + 250 = 8 ∑ i = 1 10 ( x i − 5 ) 2 = ∑ ( x i 2 − 10 x i + 25 ) = ∑ x i 2 − 10 ∑ x i + 250 = 8
Substituting ∑ x i = 50 \sum x_i = 50 ∑ x i = 50 :
∑ x i 2 − 10 × 50 + 250 = 8 ⟹ ∑ x i 2 − 500 + 250 = 8 ⟹ ∑ x i 2 = 258 (Equation 4) \sum x_i^2 - 10 \times 50 + 250 = 8 \implies \sum x_i^2 - 500 + 250 = 8 \implies \sum x_i^2 = 258 \quad \text{(Equation 4)} ∑ x i 2 − 10 × 50 + 250 = 8 ⟹ ∑ x i 2 − 500 + 250 = 8 ⟹ ∑ x i 2 = 258 (Equation 4)
Given ∑ i = 1 10 ( x i − β ) 2 = 98 \sum_{i=1}^{10} (x_i - \beta)^2 = 98 ∑ i = 1 10 ( x i − β ) 2 = 98 :
∑ ( x i 2 − 2 β x i + β 2 ) = ∑ x i 2 − 2 β ∑ x i + 10 β 2 = 98 \sum (x_i^2 - 2\beta x_i + \beta^2) = \sum x_i^2 - 2\beta \sum x_i + 10\beta^2 = 98 ∑ ( x i 2 − 2 β x i + β 2 ) = ∑ x i 2 − 2 β ∑ x i + 10 β 2 = 98
Substituting Equations 2 and 4:
258 − 2 β × 50 + 10 β 2 = 98 ⟹ 258 − 100 β + 10 β 2 = 98 258 - 2\beta \times 50 + 10\beta^2 = 98 \implies 258 - 100\beta + 10\beta^2 = 98 258 − 2 β × 50 + 10 β 2 = 98 ⟹ 258 − 100 β + 10 β 2 = 98
Rearranging:
10 β 2 − 100 β + 160 = 0 ⟹ β 2 − 10 β + 16 = 0 10\beta^2 - 100\beta + 160 = 0 \implies \beta^2 - 10\beta + 16 = 0 10 β 2 − 100 β + 160 = 0 ⟹ β 2 − 10 β + 16 = 0
Solving the quadratic equation:
Discriminant D = ( − 10 ) 2 − 4 × 1 × 16 = 100 − 64 = 36 D = (-10)^2 - 4 \times 1 \times 16 = 100 - 64 = 36 D = ( − 10 ) 2 − 4 × 1 × 16 = 100 − 64 = 36
β = 10 ± 36 2 = 10 ± 6 2 \beta = \frac{10 \pm \sqrt{36}}{2} = \frac{10 \pm 6}{2} β = 2 10 ± 36 = 2 10 ± 6
So β = 8 \beta = 8 β = 8 or β = 2 \beta = 2 β = 2 . Given β > 2 \beta > 2 β > 2 , we have β = 8 \beta = 8 β = 8 .
Now, define the new data set:
y i = 2 ( x i − 1 ) + 4 β = 2 x i − 2 + 4 β y_i = 2(x_i - 1) + 4\beta = 2x_i - 2 + 4\beta y i = 2 ( x i − 1 ) + 4 β = 2 x i − 2 + 4 β
Substituting β = 8 \beta = 8 β = 8 :
y i = 2 x i − 2 + 4 × 8 = 2 x i + 30 y_i = 2x_i - 2 + 4 \times 8 = 2x_i + 30 y i = 2 x i − 2 + 4 × 8 = 2 x i + 30
So y i = 2 x i + 30 y_i = 2x_i + 30 y i = 2 x i + 30 .
For a linear transformation y i = a x i + b y_i = a x_i + b y i = a x i + b , the new mean μ = a x ˉ + b \mu = a \bar{x} + b μ = a x ˉ + b and new variance σ 2 = a 2 × ( original variance ) \sigma^2 = a^2 \times (\text{original variance}) σ 2 = a 2 × ( original variance ) .
Here, a = 2 a = 2 a = 2 , b = 30 b = 30 b = 30 , x ˉ = 5 \bar{x} = 5 x ˉ = 5 , original variance = 4 5 = \frac{4}{5} = 5 4 .
Thus:
μ = 2 × 5 + 30 = 10 + 30 = 40 \mu = 2 \times 5 + 30 = 10 + 30 = 40 μ = 2 × 5 + 30 = 10 + 30 = 40
σ 2 = ( 2 ) 2 × 4 5 = 4 × 4 5 = 16 5 \sigma^2 = (2)^2 \times \frac{4}{5} = 4 \times \frac{4}{5} = \frac{16}{5} σ 2 = ( 2 ) 2 × 5 4 = 4 × 5 4 = 5 16
Now compute:
β μ σ 2 = 8 × 40 16 5 = 320 16 5 = 320 × 5 16 = 1600 16 = 100 \frac{\beta \mu}{\sigma^2} = \frac{8 \times 40}{\frac{16}{5}} = \frac{320}{\frac{16}{5}} = 320 \times \frac{5}{16} = \frac{1600}{16} = 100 σ 2 β μ = 5 16 8 × 40 = 5 16 320 = 320 × 16 5 = 16 1600 = 100
Therefore, β μ σ 2 = 100 \frac{\beta \mu}{\sigma^2} = 100 σ 2 β μ = 100 .